[molpro-user] Molpro and GAMESS Results Differ: CASSCF(10,10) Calculation

Peter Knowles KnowlesPJ at Cardiff.ac.uk
Mon Feb 25 06:18:01 GMT 2008


I do not think there is a problem here, or anything that needs  
resolving, or, indeed, anything that can not be understood in a  
straightforward way.

1. The singlet B_1 state must of course be much higher in energy; in  
the simplest model, its energy is 2K greater than that of the triplet,  
where K=(ab|ba) is the exchange integral for the two singly-occupied  
orbitals a, b.

2. The singlet A_1 state can of course have a lower energy because of  
configuration interaction involving the configurations a^2 and b^2.  
There is some discussion of this at, for example, http://www.ncsu.edu/chemistry/das/hundpauli.pdf 
  , although for the case where a and b have the same symmetry, where  
the details differ slightly.

3. {mcscf;.....;wf,30,1,0;......} with your input gives  -154.909207034

Peter

On 20 Feb 2008, at 08:40, rossi at york.cuny.edu wrote:

> Hello:
>
> Can someone please help me with the following problem?
>
> I cannot get GAMESS  and Molpro to give the same results.  It must be
> something that I am overlooking.
>
> 1) When the the central bridgehead bond in bicyclobutane is  
> stretched, a
> diradical is obtained.
>
> 2) The molecule has C2V symmetry, and the diradical is in a B_1 state.
>
> 3) The calculation is CASSCF(10,10).
>
> Here are the results:
>
>                  GAMESS                        Molpro
> -----------------------------------------------------------------------
> Triplet        -154.91044949                -154.91044949
> Singlet        -154.90920703                -154.72707529
>
> Why does the Molpro calculation for the singlet diradical not  
> agree.  The
> orbitals for the CASSCF are the same in both cases.  I checked this  
> over
> and over again.
>
> The input for the triplet CASSCF and triplet SCF (source of starting
> orbitals) is given below.
>
> Thank you so much for helping me ith this.  What am I missing or
> overlooking?
>
> Regards,
>
> Angelo
>
> ***,bicyclobutane - diradical
> memory,128,m
> FILE,2,bcb_1b1_nory2.wfu,old
> print,basis,orbitals
>
> basis=6-31G*
>
> geomtyp=xyz
> geometry
>  10
> Bicyclobutane - Diradical
> C1      -1.0619664679   0.0000000000  -0.1207636814
> C2       1.0619664679   0.0000000000  -0.1207636814
> C3       0.0000000000   1.1269480795  -0.0595853066
> C4       0.0000000000  -1.1269480795  -0.0595853066
> H1      -2.0727315725   0.0000000000   0.2423446764
> H2       2.0727315725   0.0000000000   0.2423446764
> H3       0.0000000000   1.6996903596   0.8655887684
> H4       0.0000000000  -1.6996903596   0.8655887684
> H5       0.0000000000   1.8197114977  -0.8963535723
> H6       0.0000000000  -1.8197114977  -0.8963535723
> end
>
> {casscf
> orbital,2140.2
> closed,5,2,3,0
>   occ,8,5,5,2
>    wf,30,2,2
>  state,1; weight,1,0;
>  orbprint,20
> }
>
> put,molden,bcb_3B1.molden;
> ---
>
> Starting orbitals can be obtained from the following input.
>
> ***,bicyclobutane - scf triplet
> memory,128,m
> FILE,2,bcb_3b1_scf.wfu,new
>
> print,basis,orbitals
>
> basis=6-31G*
>
> geomtyp=xyz
> geometry
>  10
> Bicyclobutane - Ground State
> C1      -1.0619664679   0.0000000000  -0.1207636814
> C2       1.0619664679   0.0000000000  -0.1207636814
> C3       0.0000000000   1.1269480795  -0.0595853066
> C4       0.0000000000  -1.1269480795  -0.0595853066
> H1      -2.0727315725   0.0000000000   0.2423446764
> H2       2.0727315725   0.0000000000   0.2423446764
> H3       0.0000000000   1.6996903596   0.8655887684
> H4       0.0000000000  -1.6996903596   0.8655887684
> H5       0.0000000000   1.8197114977  -0.8963535723
> H6       0.0000000000  -1.8197114977  -0.8963535723
> end
>
> {hf
> wf,30,2,2}
>
> put,molden,bcb_3b1_scf.molden;

--
Prof. Peter J. Knowles
School of Chemistry, Cardiff University, Main Building, Park Place,  
Cardiff CF10 3AT, UK
Telephone: +44 29208 79182 Fax: +44 2920874030 Email KnowlesPJ at Cardiff.ac.uk
WWW http://www.cardiff.ac.uk/chemy/contactsandpeople/academicstaff/knowles.html





More information about the Molpro-user mailing list